|
|
|
Cagliari, Italy 15-18 September 2003 ALT V Conference
Michael Cysouw
Towards a typology of pronominal
cliticization Over the last decades, there has been a wealth of publications on pronominal clitics. However, the vast majority of these works discuss language-specific characteristics only. In this talk I will present some cross-linguistic generalisations on the distribution of pronominal clitics. As a first attempt, the following typology has been made:
A) Simple cliticization: The placement of clitics is identical to the placement of the corresponding full form. B) Wackernagel cliticization: The clitics occur on second position - although the definition of this position varies somewhat between languages. C) Predicate cliticization: If clitics are present, they will attach to a particular lexical class (i.e. verbs) - however, they do not obligatorily appear. D) Apparently free cliticization: The clitics appear in various places in the sentence.
This typology turns out to be too coarse-grained. In fact, most languages that have pronominal clitics show some variability in the placement of their clitics and are thus not unequivocally applicable to one of the four major types. In my present research, I am investigation which factors influence the placement of pronominal clitics in a typological sample of the world's languages. Currently, I include the following factors in my research, all of which influence the placement of pronominal clitics in at least some of the languages in the sample:
- presence of negation - presence of auxiliaries - presence of modal adverbs - indefinite pronouns subject - questions - imperatives/hortatives/purposive/irrealis mood - information peak ('focus')
The basic process appear the some kind of 'attraction', i.e. the pronominal clitic is attracted to a specific element in the clause, for example a negative particle or a WH-pronoun. As a preliminary generalisation, this 'attractions' will be analysed as a movement of the clitic towards the information peak ('focus') of the sentence.
|